SE465342B - THREE-PHASE VOLTAGE DRIVE CONVERTER INCLUDING TWO SEX-PULSE SUB-CURRENT CONVERTERS - Google Patents

Abstract

A three-phase voltage stiff converter (SR) has two six-pulse converters (SR1, SR2). The alternating voltage terminals of the partial converters are connected to a transformer (TR) with the aid of which the resultant alternating voltage of the converter is formed as the difference between the alternating voltages of the partial converters. Each partial converter has a separate direct voltage source (C1, C2). The converter has control members (SD) which control the two partial converters with a mutual phase displacement alternating between +150 DEG and -150 DEG .

Description

Ps. i4e5 342 Another way to achieve an improved curve shape is to increase the converter's pulse rate, ie the number of commutations per AC voltage period. A three-phase converter in its simplest form has the pulse number 6. A significant improvement of the curve shape can be achieved by doubling the converter's pulse number. This can be achieved in a known manner through so-called twelve-pulse connections, where the voltages from two phase-shifted six-pulse converters are combined. There are two main types of twelve pulse connections. In the case of a first known connection of this kind, two separate transformers are required and the equipment therefore becomes complicated and expensive. In the second known converter connection of this kind, only one transformer is used, but the connection has the disadvantage that circulating current will flow between the two six-pulse converters.

Special measures in the form of increased leakage reactance of the transformer or in the form of separate inductors are required to limit these circulation currents, which causes an increased complication of the converter device.

The circulation currents, which cannot be completely eliminated, further give rise to extra losses in the inverters and require an oversizing of them.

A further way to achieve an improved curve shape is the so-called double six-pulse coupling. This uses a single transformer with an open winding for connection to two six-pulse converters with a common DC voltage source. By an open winding is meant a winding, where both ends of each phase lead are accessible for connection from the outside.

Each phase winding has one end connected to an alternating voltage socket of one partial inverter and its other end connected to the corresponding alternating voltage socket of the other partial inverter. One can consider the three phases in one end of the now mentioned winding system as constituting a three-phase system and the three phases in the other end of the winding system as constituting a second three-phase system. The difference between these two voltage systems is the voltage system pressed on the transformer. However, this known connection has the disadvantage that the transformer is subjected to a zero-sequence voltage, which will then be supplied to the alternating voltage network to which the converter 1 / converter system is connected. This zero-sequence voltage can be limited by providing the converter system with a zero-conductivity inductor, which absorbs the zero-sequence voltage. Such an inductor, however, entails a considerable complication and increase in the cost of the plant. According to another method, the transformer can be provided with an extra unwound leg, whereby the transformer will function as a zero-sequence inductor. However, this causes a not insignificant complication of the transformer. Furthermore, in the case of such a transformer, the transformer winding connected to the mains cannot be earthed at its star point, which means that the winding must be fully insulated and thus entails an increase in the cost of the transformer.

A so-called double six-pulse connection of one of the two above-mentioned types is shown in Fig. 1. The two six-pulse converters SBA and SRB are connected on their DC sides to a common DC source, which consists of a capacitor battery C. The AC sides of the converters are via a transformer TR and inductors LA, LB, LC connected to a three-phase network N.

The converters are controlled so that their AC voltages are approximately 1500 phase-shifted. The transformer has an extra unwound leg XL.

A double six-pulse coupling of the second of the two above-mentioned types is shown in Fig. 2. It has a zero-sequence inductor LC with three phase windings on a common core.

A third known way to obtain twelve-pulse function and thus improved curve shape is a converter of the so-called NPC type (NPC = Neutral Point Clamped). However, such a converter has a more complicated main circuit and contains more semiconductor components than a traditional double six-pulse or twelve-pulse connection. In such a converter, some of the semiconductor components are further loaded harder than the others, which for a given given load requires an oversizing of the first-mentioned semiconductor components and thus an increase in the cost of the converter or a reduction of its maximum power.

In all these known cases, the harmonic reduction is thus achieved at the cost of increased losses and / or considerable complications of either the transformer or the converter.

DISCLOSURE OF THE INVENTION The invention aims to provide a converter of the type indicated in the introduction, in which - simple six-pulse bridges can be used, - a conventional simple three-legged transformer can be used, - the star point of the transformer's mains winding can be grounded, which reduces transformer insulation requirements. 542 - a good curve shape can be obtained with low commutation frequency and thus low converter losses.

What characterizes a converter according to the invention is stated in the appended claims.

The invention will be described in more detail in the following in connection with the attached figures 1-8. Fig. 1 and Fig. 2 show two previously known so-called double six-pulse connections. Fig. 3 shows an embodiment of a converter according to the invention, for phase compensation of an alternating voltage network. Fig. 4 schematically shows the structure of the main circuit of one of the two sub-inverters in Fig. 3. Figs. 5a and 5b show in block diagram form the construction of the control system at the converter according to Fig. 1. Fig. 6 and Fig. 7 show the waveforms of some of the the electrical quantities present in the converter according to Fig. 1. Fig. 8 shows how the control system shown in Figs. 5a and 5b can be modified to give the function shown in Fig. 7. Fig. 9 schematically shows how a converter according to the invention can alternatively be used to equalize peak loads in an alternating voltage network through energy storage.

DESCRIPTION OF EMBODIMENTS Fig. 3 shows a converter SR according to the invention, which is connected via inductors LA, LB, LC to a three-phase AC power network N. The converter is intended to be used for phase compensation of the network, by which is meant that the converter can generate or consume a variable reactive power, eg for increasing the power factor of the mains or for controlling the voltage of the mains. The inductors just mentioned are in series with the leakage reactances of the converter transformer and can be eliminated if the latter are large enough. The converter consists of two self-commutated voltage-rigid converters SRI and SR2, which consist of six-pulse bridges, a transformer TR and a control device SD. Each partial converter has alternating voltage sockets a1, bl, cl and a2, b2, c2 respectively. A first capacitor battery C1 is connected to the sub-inverter SR1's DC1 sockets Dlu and Dln, and a second capacitor battery C2 is connected to the sub-converter SR2's DC voltage sockets D2u and D2n. The capacitors C1 and C2 are kept charged to the appropriate voltage in the manner described below and constitute separate and galvanically different DC voltage sources, one for each sub-converter. The transformer TR has a converter winding with the three phase windings AS, BS and CS and a 465 342 star-connected mains winding with the three-phase windings AN, BN and CN. The transformer has a socket GC connected to the mains winding of the mains winding for earthing of the mains winding and connections A, B, C for connection to the mains N. Each of the three phase windings AS, BS, CS is connected at one end to an alternating voltage socket in one inverter and with its other end to the corresponding AC outlet in the other inverter.

The converter has a control system SD for controlling and regulating the function of the converter. In Fig. 3 only the control pulse devices SD1 and SD2 are shown, which emit control pulses sal, sbl, sc1 to the converter SR1 and sa2, sb2, sc2 to the converter SR2.

Fig. Ä shows the construction of the partial inverter SRI. The sub-converter SR2 is constructed in an identical way. Each phase of the converter has two series-connected semiconductor valves, such as Tau and Tan. As shown in the figure, these can be built up of extinguishable thyristors (so-called GTO thyristors).

Alternatively, the valves can be built up of non-extinguishable thyristors, which are then provided with extinguishing circuits, or possibly of transistors or other controllable valves. Each valve can consist of a number of semiconductor elements connected in series. In the antiparallel with each valve, a bypass diode, Dau, Dan, is arranged. The diode can be integrated with the controllable semiconductor element. The connection point of the valves and diodes is connected to the AC voltage socket al. The other two phases are structured in the same way. The two valves in the same phase are controlled so that one valve is always conductive and the other valve non-conductive. The three phases are controlled with a 1200 phase shift between them, whereby a three-phase voltage system will be generated on the AC output terminal a1, bl, cl.

Each inverter generates a six-pulse three-phase voltage system. The difference between these two systems occurs across the converter winding of the transformer. If the substers are in phase, the difference voltage and thus the transformer voltage will be zero. If the inverters are controlled so that they operate with 1800 phase shift, a six-pulse voltage with twice the amplitude of each inverter occurs across the transformer. According to the invention, however, the partial current converters are controlled so that their phase shift is + 150 ° or -1500, whereby a twelve-pulse voltage with a low harmonic content occurs across the transformer. f4e5 542 The now mentioned phase shift between the two sub-inverters means that at least one of the sub-inverters must work with a phase shift relative to the mains which deviates from the values of the phase shift, 0O and 1800, at which the flow of active power between the mains and the sub-converter is zero. This means that the capacitor battery of at least one inverter will be continuously charged or discharged, which would make stationary operation impossible. This problem is solved at the converter according to the invention by constantly changing the sign of the phase shift between the sub-converters during operation.

Fig. 5a shows the control and regulation system for the converter according to Fig. 1.

Each sub-converter has a control pulse device, SD1 and SD2, respectively, which in a known manner emit control pulses to the inverters. The mains voltage is sensed by means of a voltage measuring device UAM, for example a voltage transformer, and is supplied with a synchronizing device SY which emits a synchronizing signal mn which constitutes a measure of the phase position of the mains voltage, and which in the absence of further control signals from the control system relative to the mains voltage that partly the inverters operate in opposite phase and partly the AC voltage of the converter (the difference between the voltages of the inverters) is in phase with the mains AC voltage.

The voltage un from the voltage measuring device UAM is supplied to a reactive power generating means QM. For this purpose, a signal corresponding to the converter current and which is obtained by means of a current measuring device IM arranged between the converter and the mains is also input. The measuring device QM forms a signal Q which is proportional to the reactive power flowing between the converter and the mains and which is applied to a summator SM1. A reference value Qr for the reactive power is also supplied to the summer, which is obtained with the aid of a reference value sensor QP. The difference between the signals QR and Q is applied to a reactive power regulator QR with PI characteristics. The output signal uDr from the regulator constitutes a reference value for the direct voltage of the two capacitors C1 and C2. The capacitor voltages are sensed by means of voltage measuring devices UDMI and UDM2 and are supplied to summators SM2 and SM3, where they are compared with the reference value uDr. The difference signals obtained with the help of the summers are supplied with direct voltage regulators UDR1 and UDR2 with P-characteristics. The output signals 910 and 920 of the controllers are applied to summers SM4 * 465 342 and SM5, the output signals 91 and ç2 of which are applied to the two control pulse devices SD1 and SD2. When the signals p1 and 92 are zero, the two inverters operate in opposite phase, ie each commutation in one inverter is 1800 phase-shifted in relation to the corresponding commutation in the other inverter. Furthermore, in each sub-converter, in this case the three phases are 1200 phase-shifted between them, and the two valves in the same phase are 1800-phase-shifted. The phase mode is further such that the resulting AC voltage of the inverter is in phase with the mains AC voltage, so no active power flows between the mains and the inverter. Since the converters are in opposite phase, as mentioned above, the resulting alternating voltage is six-pulse, ie has a relatively high harmonic content.

A positive value of the signal 91 or 92 phases (advances) the control pulses from the corresponding control pulse devices and thus the commutations in and the alternating voltage from the corresponding partial inverters. Correspondingly, a negative value of the signal 91 or 92 delays (phase retards) the control pulses from the corresponding control pulse device.

A quantity odiff, which corresponds to a phase shift of 150, is supplied to a switchable character inverter SA. The output signal from this is applied to the summator SM4 with a positive sign and the summator SM5 with a negative sign. If the output signal from the inverter is positive, then the commutations in the converter SR1 150 are brought forward, and the commutations in the inverter SR2 are delayed 150. The opposite is the case if the output signal of the inverter has a negative sign. The control pulses from the control pulse devices SD1 and SD2 are applied to a logic network LN, which emits an output pulse when the outputs of the two sub-converters constitute the inverse of each other. The output pulses from the network LN are supplied to the character inverter SA, which is switched at each output pulse from the network LN.

The logic network LN in Fig. 5a is constructed in the manner shown in Fig. 5b.

It includes three exclusive OR gates OG1, OG2, OG3 and an AND gate AG. Each OR gate emits an output signal if its two input signals are different from each other. Each positive edge of the output signal from the AND gate AG switches the character inverter SA. If the control signals sal and sa2 are different, an output signal is obtained from the gate 0G1, if sbl and sb2 are different from each other, an output signal from OG2 is obtained, and if sc1 and sc2 are different, an output signal is obtained from OG3. A control vector for the sub-converter SR1 is defined by the quantities sal, sbl, sc1, and a control vector for the converter SR2 of the quantities sa2, sb2, sc2. One control vector is the inverse of the other if sal v sa2 ggh sbl + sb2 and sc1 + sc2, and when so, outputs are obtained from all three OR gates and an output signal from the 0CH gate AG to the character inverter to provide a character inversion.

In the steady state, the reactive power Q corresponds to the desired Qr. The output signal from the reactive power regulator QR is therefore constant and equal to the value of the capacitor voltages which gives the desired reactive power. Furthermore, the two capacitor voltages are equal to the reference value uDr obtained from the reactive power regulator, and the signals 910 and 920 are both zero. At a certain moment, the output signal from the character inverter SA is positive. The next commutation will therefore be preceded by 150 in the converter SR1 and postponed by 150 in the converter SR2, ie the commutations will take place with a 30 ° phase shift. In this commutation, the character inverter is switched, and pl and 92 change characters.

The next commutation will therefore be delayed by 150 in the converter SR1 and earlier by 150 in the converter SR2. In this way, the two partial inverters will work with a mutual phase shift that is alternately + 150 ° and -1500. A commutation will take place in one of the inverters every 30 ° and the resulting AC voltage of the inverter will be twelve-pulse with a low harmonic content. Each sub-converter will alternately be 15 ° phase-advanced and respectively 150 phase-decelerated in relation to the phase position at which the active power flow is zero. This means that power will alternately flow out of the converter's capacitor and into the same capacitor. With the control principle now described, however, the average value of this current will be zero, ie the capacitor charge will be constant.

The six uppermost curves in Fig. 6 show the control pulses to the partial inverters.

When a control pulse is "1", the upper valve in the corresponding phase is conductive, and when a control pulse is "0", the lower valve in the phase is conductive. As an example, the phase voltages ual and ua2 of the two sub-inverters are shown below in Fig. 6. The resulting alternating voltage belonging to phase A is shown below, which constitutes the difference between the voltages ual and ua2. At the bottom of Fig. 6, the two currents floating to the capacitor batteries in iD1 and iD2 are shown. oh! 9 465 s42g The closed control circuit formed by the units UDM1, SM2, UDR1, SD1 and SRI will by keeping the phase position of the control pulses to the converter SR1 and thus by the flow of active power and the capacitor current iD1 continuously keep the voltage of the capacitor C1 equal. as the reference value uDr. Correspondingly, the voltage of capacitor C2 will be kept equal to the same reference value. The parent circuit for controlling the reactive power includes the units QM, QP, SM1 and QR. Thus, for example, an increase of the reference value QR will give rise to an increase of the voltage reference uDr and thus in the manner described above an increase of the capacitor voltages and thus of the converter's total output amplitude until the converter's flow of reactive power Q corresponds to the reference value QR.

The interventions caused by the control circuits now mentioned in the control of the inverters are preferably small and do not affect the previously described principal mode of operation of the inverter.

In the above-described embodiment of the invention, as can be seen from Fig. 6, the joint intervals of the upper semiconductor valve in each phase are shorter than the joint intervals of the lower valve in the same phase. The latter valve will therefore have a stronger thermal load, which reduces the maximum power of the inverter. The start state of the character inverter SA determines which of the valves have longer and shorter joint intervals, respectively, ie in which of two possible modes the system operates.

In one embodiment of the invention, a change of the mode in which the converters operate is performed periodically. This can be effected by controlling the converters in accordance with the diagrams in Fig. 7. In short, the control strategy means that one of the recurring character changes in SA is inhibited. In Fig. 7, the system operates in a first mode before t = tl. At t = t1 a mode change is initiated, which is completed at t = t2, after which the system operates in its second mode.

By periodically changing the operating mode in this way, each individual inverter valve will belong to the valve group that works with long joint intervals and to the one that works with short joint intervals.

If the period time for the mode change carried out in this way is short or comparable to the thermal time constant of the valve, the temperature rise caused by the losses will be the same in all the valves of the converter. These can then be utilized in an economically optimal way. 465 342 '° The mode change can take place as often as with every other commutation, ie the character reversal sequence then becomes ++ - ++ - ++ - ....

Fig. 8 shows a device for carrying out the mode change in the manner described here. The device comprises an additional character inverter SB in the form of a multiplier, which character inverts the output signal from the character inverter SA (cf. Fig. 5b). The multiplier is controlled by the output of a bistable circuit BC. This has a preparatory D input which is controlled by an oscillator OSC which produces a square wave with a lower frequency than the commutation frequency of the converters. At the commutation that follows immediately after each state change of the output signal of the oscillator, the output signal of the circuit BC will change characters, which causes the system to change operating mode.

Instead of the free-swinging oscillator shown in Fig. 8, the mode shift can be controlled in another way, for example depending on some measured system quantity, for example the temperature of the semiconductor elements.

Fig. 9 shows an alternative use of a converter according to the invention. The DC voltage sources of the two sub-converters in this case consist of accumulator batteries B1 and B2. At low load in the mains N, the converter is controlled so that active power flows from the mains into the accumulator batteries where it is stored. At high load, the inverter is controlled so that active power flows from the batteries to the mains. In this way, in a manner known per se, an equalization of the loads of the network can be made. The control and function of the converter can in principle be the same as described above in connection with Figs. 3-8, whereby of course the control circuits for controlling the individual capacitor voltages are replaced by control circuits which control the current through each accumulator battery. The individual control then ensures that the average current through each battery is equal to the common reference obtained from the overall control.

Likewise, the control circuit shown in Fig. 5 is suitably supplemented with means for controlling the active power flow by common phase advancement and phase deceleration of both partial inverters.

The embodiments and uses described above are only examples and both other embodiments and areas of use are conceivable within the scope of the invention. The principle according to the invention is thus described for six-pulse converters whose output voltages are combined so that a twelve-pulse voltage is obtained, but the principle according to the invention is also applicable to other pulse numbers than those mentioned above. The value of the signal described above

Claims (12)

H 465 542 çdiff, which gives a mutual phase shift of 1500 between the two sub-inverters, gives the lowest harmonic content, but other values of the phase shift between the two sub-inverters can also be used. In a manner known per se, in the case of a converter according to the invention, one or more additional commutations can be made every half period in order to further reduce the harmonic content of the output voltage. The control procedures described above are examples only. As can be seen from the above description, according to the invention a converter is obtained where with a low commutation frequency and thus low losses a low harmonic content can be obtained. No zero-voltage voltages or circulating currents occur and therefore no special inductors or special transformer designs are needed to eliminate such problems. Both the sub-inverters' main circuits and the transformer have the simplest and most economical structure possible. A converter according to the invention therefore becomes simple and economically advantageous. PATENT REQUIREMENTS A three-phase voltage-rigid converter (SR), comprising a first (SR1) and a second (SR2) six-pulse converter, each sub-converter with alternating voltage socket (a1, bl, cl; a2, b2, c2) and DC socket (Dlu, Din; D2u, D2n) at which the inverters operate phase-shifted in relation to each other, the AC outlets of the inverters are connected to a transformer (TR) to generate the AC voltage of the inverter by combining the AC voltages of the inverters, and the DC sockets of each inverter are connected to a DC voltage C1 ), characterized in that the inverter has two separate DC voltage sources (C1, C2), each connected to the DC sockets of one of the inverters, and that the inverter is provided with the control means (SD) arranged to control the inverters so that the phase shift ( çdiff) between the two sub-inverters constantly alternating between positive and negative values. 2. Converters according to claim 1, characterized in that the control means (SD) are arranged to control the partial inverters so that the average value of the phase shift (çdiff) between them is zero. Il 465 342 3. Converters according to any one of the preceding claims, characterized in that the control means are arranged to control the partial inverters so that the phase shift (çdiff) between them alternates between a positive value and an equally large negative value. 4. Converters according to any one of the preceding claims, characterized in that the inverters are six-pulse and that the control means are arranged to control the inverters so that the phase shift (çdiff) between them alternates between two values which are substantially equal to + 15o ° and -15 °. Converters according to any one of the preceding claims, characterized in that the control means are arranged to control the partial inverters so that the phase shift (pdiff) between them periodically alternates between said two values. A converter according to claim 5, characterized in that the control means are arranged to control the sub-converters so that the phase shift between them changes characters each time the control vector in one sub-converter is the inverse of the control vector in the other sub-converter. 7. A converter according to claim 6, characterized in that the control means comprise means (OSC, BC) for suppressing a single character change to effect a change of the operating mode of the system. Converter according to claim 7, characterized in that the control means comprise means (OSC) for suppressing individual character changes with a frequency which is lower than the commutation frequency of each subst. Converter according to one of the preceding claims for connection to an alternating voltage network and for emitting a positive or negative reactive power to the network, characterized in that the direct voltage sources consist of capacitor batteries (C1, C2). Converter according to Claim 9, characterized in that it is provided with means for controlling the capacitor voltages (uD1, uD2) in dependence on the flow of reactive power (Q) between the current-rich fi arê fi 0Ch mains by influencing the phase position of the sub-converters relative to the mains voltage. ”465 342 Converter according to one of the preceding claims, characterized in that the transformer (TR) has an open converter winding with three phase windings (AS, BS, CS), each phase winding having its one end point connected to an alternating voltage socket (t ex a1) of one sub-converter (SR1) and its other end point connected to the corresponding AC voltage socket (a2) of the other sub-converter (SR2). Converter according to claim 11, characterized in that the transformer (TR) has a star-connected mains winding (AN, BN, CN) with a socket (GC) for earthing the winding.